Histopathological image analysis

A non-transitory computer-readable medium having embodied thereon a computer program comprising computer program code that when executed causes the computer to:

• obtain (210), by the computer, a plurality of first microscopic images of first histological specimens that have been stained with a first marker;
• obtain (212), by the computer, a respective plurality of second microscopic images of second histological specimens that have been stained with a second marker, different from the first marker, each one of the second histological specimens corresponding to a respective one of the first histological specimens;
• obtain (220), by the computer, a respective plurality of mask images generated for the second microscopic images, each mask image identifying a histological feature of interest highlighted in the respective second microscopic image by the second marker;
• for each first microscopic image and the respective second microscopic image and mask image:
• obtain (230, 270), by the computer, a transformation that maps between the first microscopic image and the second microscopic image, and
• align (232, 272), by the computer, the first microscopic image and the mask image, using the transformation;
• and
• train (240), by the computer, the machine-learning algorithm with the plurality of aligned first microscopic images and mask images, and
• thereby train the machine-learning algorithm to predict, from another first microscopic image of another histological specimen stained with the first marker, a histological feature of interest that would be highlighted in that specimen by the second marker.